GitHub - ShanikaCPerera/beacon: A deep learning framework that uses large language models (LLMs) to generate dense, contextual embeddings from raw sandbox-generated behavior reports. These embeddings capture semantic and structural patterns of each sample and are processed by a 1D CNN network for multi-class malware classification.

Name		Name	Last commit message	Last commit date
.gitignore		.gitignore
bosansky_f1_scores.py		bosansky_f1_scores.py
chroma_db_manager.py		chroma_db_manager.py
config.py		config.py
data_preparation.py		data_preparation.py
data_preprocessing.ipynb		data_preprocessing.ipynb
embedding_client.py		embedding_client.py
json_chunker.py		json_chunker.py
json_extraction.py		json_extraction.py
logfile_chunking.py		logfile_chunking.py
logfile_chunking_embedding.py		logfile_chunking_embedding.py
logfile_embedding.py		logfile_embedding.py
requirements.txt		requirements.txt
test_num_tokens.py		test_num_tokens.py
training_CNN.ipynb		training_CNN.ipynb
training_LSTM.ipynb		training_LSTM.ipynb
training_MLP.ipynb		training_MLP.ipynb
training_SVM_Linear.ipynb		training_SVM_Linear.ipynb
training_SVM_RBF.ipynb		training_SVM_RBF.ipynb
visualization_pca.ipynb		visualization_pca.ipynb
visualization_tsne.ipynb		visualization_tsne.ipynb

About

A deep learning framework that uses large language models (LLMs) to generate dense, contextual embeddings from raw sandbox-generated behavior reports. These embeddings capture semantic and structural patterns of each sample and are processed by a 1D CNN network for multi-class malware classification.

contextual-embeddings llm malwareanalysis

Activity

1 star

1 watching

0 forks

Report repository